"Lambda" (value of fep-lambdas) is output to log at evey step for every simulation
"Lambda", which seems to be the current value of fep-lambdas, is output to log at every step of every simulation, even in non-free energy simulations and energy minimizations (next to "Time" and "Step"). In the non-free energy case it just ouputs 0.00000 at every step, and even in free energy simulations where mdp options such as "coul-lambdas" and "vdw-lambdas" it just outputs zero if "fep-lambdas" is not used. Additionally in regular free energy simulations the lambda values are constant throughout the simulation, so it would be unncessary to print it every step.
In expanded ensemble simulations it might make sense to print the value, except that it only prints the corresponding value from "fep-lambdas", not "coul-lambdas", "vdw-lambdas", etc. Additionally the expanded ensemble sims output more detail to the log that includes the current lambda values.
#3 Updated by Mark Abraham over 5 years ago
Inasmuch as (hopefully better versions of) lambda values are written to the .edr file, there's no need to have it in the .log file.
If there's value in observing any transitions of the lambda value of the simulation reported in the .log file, then that transition event is reasonable to log independently. But frankly, I can't see that that is useful for more than debugging the .mdp input (except expanded ensemble, which takes care of itself).
So I'm fine with removing this